AbstractConsider a millimeter‐wave (mmWave) massive multiple‐input multiple‐output (MIMO) system, where a base station (BS) relying on discrete lens antenna (DLA) array transmits scalable video code (SVC) video to multiple single‐antenna vehicles. Taking into account both the video quality for vehicles and the communication resource consumption of BS, we define a new metric given by the quality of received video to the power consumption ratio (QPR) in order to measure the system efficiency. The objective of this article is to maximize the minimum QPR among vehicles by designing the antenna selection and power allocation scheme, while ensuring the constraints of total transmit power and DLA array structure. Due to the piecewise form of the video quality function, it is intractable to solve the formulated problem, which is therefore decomposed into antenna selection and power allocation subproblems. A maximum equivalent channel gain (MECG) algorithm and an iteration algorithm are proposed for antenna selection subproblem with different complexities, respectively; while an optimal power allocation (OPA) algorithm is designed by analyzing three cases of transmit power constraints. Simulation results indicate that the proposed MECG algorithm has moderate performance with low‐complexity, while the OPA algorithm can help to achieve the optimal QPR in contrast to other algorithms.